The Way Out of Brain Fog

A silhouette with the face blurred out by a bright orange light
Yifei Fang / Getty

Debbie Gustafson of Dresher, Pennsylvania, was on the trip of a lifetime, touring the Galápagos with her family last March, when she began to feel the effects of COVID-19. Though her physical symptoms—diarrhea, dry cough, chills—were considered mild by doctors, her fatigue was crushing, and her mind was trapped in a fog. Once an avid reader, she couldn’t get through a page. “My eyes darted everywhere. I had no focus,” she told me. Before COVID-19, she’d held two part-time jobs, but she soon had to give up both of them.

The cognitive problems emerging from mild to moderate cases of COVID-19 are so new that researchers have struggled to define them. A just-published study led by Igor Koralnik, the director of the Neuro COVID-19 Clinic at Chicago’s Northwestern Memorial Hospital, analyzed the first 100 “long COVID” sufferers who came to the clinic, either in person or via virtual visits. None had ever been hospitalized for COVID-19, yet 85 percent had four or more neurological complaints, including “brain fog”—persistent trouble with focusing, retaining short-term memories, and managing complex tasks.  

In February, the National Institutes of Health gave long COVID a clinical name: Post-Acute Sequelae of SARS-CoV-2 infection (PASC). But official recognition doesn’t tell us what percentage of COVID-19 sufferers experience lingering neurocognitive problems, or how many long-COVID patients there are. “There’s no numerator or denominator for the group yet,” says Sara Manning, a neurologist at the new Post-COVID Assessment and Recovery Clinic at the University of Pennsylvania, one of dozens of such clinics springing up in the United States and worldwide. That Koralnik’s study, like many of the new clinics, probably does not include many people without the resources or connections to find their way to specialized care only increases the uncertainty. “It’s likely there are many millions of these patients in the U.S., and dozens of millions in the world,” Koralnik told me. Like Gustafson, many of them are struggling with brain fog, and with its profound and often frightening disruptions to their daily lives. Now, in the second year of the pandemic, researchers and therapists are beginning to understand how to help them.

More in this series


While identifying long-COVID patients is challenging, helping them through the uncharted territory of their illness is more difficult still. Benjamin Abramoff, the director of Penn’s post-COVID clinic, is a physiatrist with a specialty in spinal-cord injury. Like most U.S. physicians, he was abruptly introduced to COVID-19 a year ago, when hospitals were inundated with critically ill patients—some of whom suffered brain damage from oxygen loss, blood clots, or strokes. Many of those who survived with the help of ventilators emerged, expectedly, with “post-ICU syndrome,” a series of deficits that include memory, attention, and processing-speed impairments. The sickest patients can experience hallucinations and psychoses. “A pneumonia patient in the ICU can be delirious,” Abramoff says. “It’s not a surprise.”

But no one was prepared for the second wave of neurocognitive complaints, which came from people who had never been hospitalized. One man in his early 30s, whose physical COVID-19 symptoms had been limited to shortness of breath, appeared in Abramoff’s clinic in January, concerned about his ability to do his job. His work involved keeping many things in his head at once, and he could no longer manage it. Another man, a physician responsible for complex and nuanced diagnoses of heart patients, could not remember phrases that he used every day at work, such as blood pressure. He said he sometimes felt disembodied—depersonalized, as if he were outside of himself.

As these new patients—some 350 to date—arrived at the Penn clinic, Abramoff noticed the patterns now codified by Koralnik and his colleagues. For some people, coordinating Zoom calls and emails was too much to handle. Others could no longer work at all. Some began to get better after a couple of weeks or months, but a worrying minority remained ill. Abramoff had never seen so many people with a single viral illness stay so impaired for so long.

At Beth Israel Deaconess Medical Center in Boston, patients with milder COVID-19 began requesting help with mental fatigue and concentration problems late last year. The cognitive neurologist Tamara Fong thought they resembled patients with post-concussive syndrome (PCS), which some neurologists hypothesize results from inflammation of the brain. Like PCS patients, many members of the long-COVID group were brain-fogged and depressed. They often had headaches and trouble sleeping.

So far, Fong says, the most effective treatments for long COVID resemble those for physical brain injuries. She starts by restoring good sleep hygiene, limiting daytime naps and screen time before bed. She also wants her patients to reduce stress: Because heavy exercise tends to be too taxing at first, she recommends yoga, meditation, or tai chi. “Mindfulness helps,” she told me. After patients are rested and have learned to relax, she helps them recover their cognitive function through steady, gradual practice. Patients might start by reading newspaper headlines and short articles. “Doing too much too fast is like trying to run a marathon without training,” she said.

At Penn, therapists also target particular cognitive deficits through regular mental exercises, aiming to stimulate growth of neurons and strengthen their interconnections in affected areas of the brain. Alexandra Merlino, a speech-language pathologist who conducts post-COVID cognitive rehab at the clinic, might ask patients with word-retrieval problems to listen to a podcast and summarize it—concisely and in concrete terms. “Pronouns are not allowed,” she told me. She encourages patients to practice in everyday life, too, by participating more in discussions with friends.

To improve memory, Merlino might have patients remember a short grocery list. Over time, her memory-impaired COVID-19 patients can go from recalling three items to six to a dozen or more, both by strengthening the memory center of their brain and by learning to compensate for their acquired deficits. “We teach them tools like association and categorization to remember items,” she said.

Because many people with long COVID are hypersensitive to external stimuli, they must also practice their recall in real-world conditions, such as the noise and bright light of a supermarket. Sunglasses or earplugs can help them reduce disruptive stimuli at first, but Merlino tries to gradually increase their exposure. “Maybe they would bring the earplugs and sunglasses,” she said, “but for a few minutes they may try to grocery shop without them and only put them on when they start to experience symptoms.”

People can also offset their cognitive symptoms by entering appointments in reminder apps, recording important details in memory journals, and making lists of the steps required to complete a task. “Many of these people have never had memory or organization problems before,” Merlino said, “but suddenly they need to function in the here and now.”

For Debbie Gustafson, the Penn-clinic patient, therapy was staged: She needed to recover physical, emotional, and cognitive skills, in that order. Without physical and emotional reserves, addressing cognitive problems would be especially difficult. She went through a six-week “pulmonary-wellness boot camp” to strengthen breathing muscles that had been weakened by COVID-19. Group therapy led by a psychotherapist helped her realize she wasn’t alone. “One of the most important things in this process was just being heard,” she told me. Only then did she begin making simple plans—such as scheduling breakfast between 8 and 9 a.m.—in order to recover her ability to organize. Gustafson now starts each day with a detailed schedule, arranged in tiny increments so that she doesn’t tire herself. One day she might have half an hour for yard work; another day, half an hour for a walk. Reading for 10 minutes a day helps repair the part of her brain involved in memory and focus.

Merlino reports improvements for Gustafson and others at the Penn clinic. The majority of patients have noticeably improved after two months, and still more are improved after four, she told me.  

At Beth Israel, most of Fong’s patients appear to be better by the six-month mark, and many have resumed their pre-COVID responsibilities. But there are caveats. Some 50- and 60-year-olds are returning to jobs before their younger peers, perhaps because the more active, stressful lives of young parents and early-career employees can be more cognitively demanding. Because of these pressures, younger people may need more time to recuperate and regain their capacities.  

It’s too early to tell whether patients can recover completely. Merlino has seen a couple of patients with minor cognitive deficits apparently recover all their skills and capabilities. But once patients have recovered enough to use the clinic’s strategies at home, they are discharged, and the clinic can no longer keep track of them. “The brain takes a long time to heal, and there is a limit to what insurance companies will cover,” Merlino said. “A lot of my patients have returned to work, and their functioning has definitely improved. In some cases, it has been over a year and they are still not back to work.”

The sooner rehab starts, the better the outcome is likely to be, says the neurologist Michael Zandi, a co-founder of a long-COVID clinic at the National Hospital for Neurology in London. While some people may ignore mild symptoms in hopes that they will go away, or delay treatment for fear that word of their symptoms will get out and jeopardize their jobs, Zandi encourages them to seek help: “People should be aware that if they’ve got numbness or weakness, serious memory problems, this could have something to do with their brain.”  


As researchers and therapists develop treatments for the neurological symptoms of long COVID, other specialists are working to identify its causes. Many think the symptoms result from cytokines, molecules produced by the immune system in response to infection. While it’s not clear to what extent the virus can enter the brain, cytokines can cross the blood-brain barrier, and they may be provoking an inflammatory response. To test the idea, the Columbia University neuroscientist J. John Mann plans to scan patients’ brains in search of a particular protein that is activated during an inflammatory response. In patients with depression and suicidal ideation, Mann has watched levels of that protein surge; he aims to study whether the same is true for those with long COVID.

Researchers at Yale also suspect that inflammation generated by the immune system is the cause of many long-COVID symptoms. Like Koralnik, they point to evidence that people already experiencing depression or an autoimmune disease, both of which are associated with inflammation, appear to be at heightened risk of neurological complications from COVID. They have also observed that long-COVID patients have a higher-than-average incidence of psychosis, which might also be linked to inflammation. Serena Spudich, a clinician at the university’s new neuroCOVID-19 clinic, has found inflammatory proteins and antibodies in patients’ blood samples. More alarming, she and her colleagues have also found autoantibodies—immune molecules that attack the patients’ own tissues instead of the pathogen. Spudich speculates that COVID-19 might cause a subtle injury to the blood-brain barrier that allows the autoantibodies to access and attack brain tissue. If that is so, targeted immune-modulating therapies of the sort used for neurodegenerative or autoimmune diseases could help.

All of these puzzles will be easier to solve when researchers can include more, and more diverse, patients in their studies. The NeuroCOVID Project, a collaboration between the NIH and NYU, is assembling a repository of blood, tissue, and spinal-fluid samples from large numbers of patients with neurological complications from COVID-19, along with a data bank informed by their physicians. Researchers currently limited to studying small patient groups at elite clinics hope that the project will provide a much fuller picture of patients’ experiences. “Investigators will be able to access the database and the biobank to answer questions of their own priority,” says Andrea Troxel, the project coordinator and director of the biostatistics division at NYU Langone Health. For instance, a few long-haulers have reported some relief of their COVID-19 symptoms following vaccination; the resources of the NeuroCOVID Project could help researchers investigate these anecdotes.

COVID-19 infections may be physically disrupting our brains, but life in the age of the coronavirus is scrambling them, too. “In urban environments, in closed spaces, we’ve had to redefine our roles to balance our lives. We’re teaching kids in one room and Zooming all day in another—constant alternating stressors,” says Leo Shea, a senior psychologist at NYU Langone. Like the virus, these conditions can also affect attention, memory, cognition, decision-making abilities, and emotional balance.

Seeing the positive amid the losses of the pandemic could be the biggest challenge of all. “I stopped chasing my pre-COVID capacity, because it was causing me to overlook all the progress I had made,” Gustafson told me. After following the Penn clinic’s therapeutic routines day after day for months, she said she feels “like a switch has flipped. I have enough tools now to move forward without fear I will slip back.” A year after getting COVID-19, she’s been able to return to one of her part-time jobs, managing payroll and benefits for a consulting firm. “I’ve slowly worked my way back into it, but it’s different now,” she said. “If something out of the ordinary happens, I am still completely overwhelmed.” Coming back from COVID-19 “is going to be constant learning,” she said—for patients and therapists alike.


The Atlantic’s COVID-19 coverage is supported by a grant from the Chan Zuckerberg Initiative.